Phase separation processes of polymer solutions, including those of acrylonitrile, have been very useful for the preparation of porous low-density microcellular plastic foams, primarily in the form of fibers, sheets and blocks or slabs.
In U.K. Patent Specification No. 938,694, a microporous material is made by mixing a finely divided thermoplastic resin with a gel-forming solvent therefore, raising the temperature of the mixture above the gelling point thereof, decreasing the temperature to form a gel and removing the gel forming-solvent from the mixture by treatment with a solvent for the gel-forming solvent but not for the thermoplastic resin. In the example of this U.K. patent, 35 percent by volume of polyethylene resin was heated with 65 percent by volume of xylene at 140.degree. C and allowed to cool to room temperature, whereupon a gelled mass was formed. The mass was cut into sheets and the xylene was extracted with ethanol. After removal of the ethanol with water, microporous foam sheets were obtained, which had a pore size of below about 1.0 micron and a total porosity of about 65 percent, the sheets being useful as separators in a storage battery, for example.
In Young, et al., U.S. Pat. No. 4,430,451, such a process was used to produce low density foams from poly(4-methyl-1-pentene) resin and a solvent comprising bibenzyl and using, for example, methanol, to remove the bibenzyl leaving the resin in the form of a fragile, microcellular, low density foam, having a broadly disclosed pore volume of from 90 to 99 percent, and a specifically exemplified pore volume of about 94 percent. Such foams were machined into blocks for laser fusion targets.
In Castro, U.S. Pat. Nos. 4,247,498 and 4,519,909, the thermally-induced phase separation technology was employed to make microporous foams in forms ranging from films to blocks to intricate shapes. In the '909 Patent, it is stated in Col. 6, lines 34-35, that "as the solution is cooled to the desired shape, no mixing or other shear force is applied while the solution is undergoing the cooling." This strongly suggests that beads were not contemplated. Castro, '909 Patent, Col. 27-28, also discloses microporous polymers containing functional liquids. The polymers are said to have either a cellular or non-cellular structure in which the liquid is incorporated. A cellular structure is defined in Col. 7 as having a series of enclosed cells having substantially spherical shapes with pores or passageways interconnecting adjacent cells, the diameter of said cells being at least twice the diameter of said pores. Such a morphology is not ideal for adsorbing large molecules because the passageways are not of uniform diameter and this represents a serious drawback for large molecule absorption and desportion.
Stoy, U.S. Pat. No. 4,110,529, discloses spherical polyacrylonitrile beads formed by a process in which a polymer solution is dispersed in a "liquid dispersing medium that is a nonsolvent for the polymeric material and is immiscible with the solvent." The emulsion is added "with stirring into an excess of a coagulating liquid that coagulates the polymer material...and that is a non-solvent for the polymer material, is miscible with the solvent, and is immiscible with the dispersing medium." In adopting the classical method to making beads, applicants herein can, for example, form a hot emulsion of a polymer solution in mineral oil and quench the same by adding it to mineral oil at a lower temperature. Therefore, applicants do not use a "coagulating" bath which is immiscible with the polymer solution and miscible with the dispersing medium. The main drawback with the Stoy process, however, is that, even though up to or greater than 95 percent void content is obtained, as set forth in Col. 3, lines 39-41, "a non-sticky skin is formed on the surface of the droplets at the very beginning of the coagulation." Such a skin cannot be controlled by such a process and is only partially permeable, thus substantially interfering with the absorption and desorption of large molecules, and making very desirable the production of non-skinned or controllably skinned microporous beads. Additionally, as will be shown in the comparative examples hereinafter, beads made using the process disclosed in Stoy possess nonisotropic pores, with large pores concentrated in the interior and thus further contributes to their ineffectiveness in size exclusion chromographic applications and the desorption of large molecules.
Matsumoto, in U.S. Pat. No. 4,486,549 generally discloses porous fibers and filaments, but also teaches the formation of polyacrylonitrile particles having a porous structure by adding the polymer solution dropwise into an atomizer cup in Example 1 of the patent. However, beads produced in this method have low pore volume, 0.90 ml/g, as seen in Comparative Example 1A of this application; this is responsible for low capacity. The particles tend to be flattened and non-spherical, as is shown in FIG. 20, and this will cause excessive pressure drops.
Of general interest is Josefiak et al., U.S. Pat. No. 4,594,207, in which the technology is used to produce porous bodies, such as fibers, hollow filaments, tubes, tubing, rods, blocks and powdery bodies from polyolefins, poly (vinyl esters), polyamides, polyurethanes and polycarbonates. Poly (acrylonitriles) were not used. There were adjustments in total pore volume, pore size and pore walls being made by varying solvent ratios; the pore volumes exemplified are in the 75-77.5 percent range. Josefiak discloses shaping the viscous solution by methods requiring no shearing during cooling. Examples 1-5 in the Josefiak patent describe the shaping of hollow filaments by spinning the solution through a hollow filament nozzle and then cooling; and Examples 5-7 describe the forming of membranes by coating a plate glass with the solution and then cooling. It is also noticed in Josefiak, U.S. Pat. No. 4,666,607 Col. 2, line 43 to Col 3, line 14, that he teaches away from using strong shear forces during cooling. At no point in the disclosures does Josefiak contemplate the use of turbulence during cooling, thus strongly suggesting that beads were not contemplated. In contrast, in the present invention, shear is used in the solution prior to and during cooling, so as to form droplets which cool into beads. These beads surprisingly provide a high degree of separation capability in chromatographic applications, low resistance to chromatographic flow rates and excellent morphological advantages for column packing applications, such as having good compressive stength and being substantially spherical. In Zwick, Applied Polymer Symposia, No. 6,109-149, 1967, a similar method was used to prepare microporous fibers using polymer concentrations in the wet-spinning range, 10-25 percent, producing microporous structures having pore volumes of 75-90 percent.
In Coupek et al., U.S. Pat. No. 3,983,001, is described a method of isolating biologically active compounds by affinity chromatography. The compounds isolated included enzymes, coenzymes, enzyme inhibitors, antibodies, antigens, hormones, carbohydrates, lipids, peptides, and proteins as well as nucleotides, nucleic acids, and vitamins, such as Vitamin B. Among the porous carriers are mentioned polyacrylonitrile particles, but these are macroporous, require secondary shaping processes to form particles from the gel obtained by practicing this invention, and are inferior in other chromatographic processes, particularly for size exclusion chromotography.
The current state of the art of microporous beads for purification, chromatography, enzyme binding and the like, are represented by the highly porous hydrophylic resins for sale under the trademark SEPABEADS.RTM. by Mitsubishi Chemical Industries Limited. These are said to comprise spherical hard gel beads composed of highly porous hydrophilic vinyl polymer. They have an average diameter of 120 microns and a pore volume of less than 1.6 ml/g. Also to be mentioned, the same company produces DIAION.RTM. highly porous polymer beads comprised of styrene crosslinked with divinyl benzene. Such beads can have a narrow pore size distribution, their pore volume is less than 1.2 ml/g.
It is thus apparent from the state of the art set forth above that a major drawback of many microporous polymer structures has been the pore volume being less than desired, typically from 20 to 75 percent of the polymer structure, or up to 90 percent, but, as seen in Castro, mechanical strength difficulties arise. Lower void volume enhances mechanical strength, but produces low capacity when used in structures such as chromatography adsorbants. Other prior art structures are in the shape of fibers, filaments or membranes and cannot be effectively used to pack chromatographic columns, thus requiring costly secondary shaping equipment. Many of the prior art structures are not rigid and may swell with changes in ionic strength or solvent, making column packing and control difficult.
It has now been discovered that microporous beads, substantially spherical in shape, having very high void volume, a substantially skinless surface, large pore diameters and high mechanical strength can be produced in thermal-induced phase separation methods by judicious selection of process techniques. Such beads are novel and their valuable properties are entirely unexpected in view of the prior art and the best materials made commercially available to date. The substantially skinless beads of this invention permit access of large molecules to their inner surface areas. They are made by a process which does not involve difficultly-controllable chemical reactions, such as formation of porous beads from monomers. The morphology of the beads makes them ideally suited for most chromatography applications, especially for the chromatography of proteins. They can also be used for enzyme immobilization, and for many other applications.